skill-maker
Pass
Audited by Gen Agent Trust Hub on Jul 31, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is a meta-tool for skill development and does not contain any malicious code, data exfiltration patterns, or unauthorized command execution.
- [COMMAND_EXECUTION]: The skill includes a local Node.js script (
scripts/validate-skill-maker.mjs) used for linting and structural validation of the skill folder. This script only uses built-in Node.js modules (fs,path,process) to perform local file system operations and does not execute arbitrary shell commands or access the network. - [EXTERNAL_DOWNLOADS]: The skill references official documentation from trusted organizations (Anthropic and OpenAI). These URLs are used solely for informational purposes and to provide snapshots of guidance for the agent's reasoning.
- [PROMPT_INJECTION]: The skill explicitly includes rules and test cases (in
assets/evals/skill-maker-cases.jsonl) designed to teach the agent how to identify and reject prompt injection attempts in retrieved data, reinforcing safety protocols.
Audit Metadata